Lung cancer is the leading cause of cancer-related deaths worldwide. According to global cancer statistics, approximately 350 deaths occurs every day from lung cancer in 2022 (1). The pathogenesis of lung cancer is very complex, involving many factors, such as environmental factors, genetic factors, and lifestyle factors such as smoking. According to histological classification, lung cancer can generally be divided into two subtypes: small cell lung cancer and non-small cell lung cancer (NSCLC). About 15% of lung cancer cases are small cell lung cancer, while NSCLC accounts for about 85% of lung cancer cases and the majority of lung cancer morbidity and mortality (2). NSCLC can be further divided into adenocarcinoma, squamous cell carcinoma, and large cell carcinoma, among which adenocarcinoma is the most common, accounting for about 40% of all NSCLC cases (3). For patients who are diagnosed with early-stage NSCLC, the survival rate is relatively high after surgery (4). However, in many cases, patients with early-stage NSCLC do not show typical clinical symptoms, making early screening of NSCLC very difficult, resulting in more than half of NSCLC patients already developed local or distant metastases at diagnosis (5). At present, the main treatment strategy for NSCLC is still based on surgical resection, combined with adjuvant chemotherapy, radiotherapy, immunotherapy, and targeted therapy. However, only about 20-30% of patients are clinically suitable for surgical treatment (6), and the recurrence rate after surgery is relatively high. In addition, conventional chemotherapy-based regimens are associated with limited efficacies in clinical practice, and radiotherapy is associated with side effects and reduced quality of life. Due to these limitations of conventional treatment strategies, the development of novel treatment for NSCLC represents a pressing unmet need. With the development of molecular biology, cancer-targeted therapy in NSCLC has gained increased attention. Considering the significant heterogeneity of NSCLC, treatment regimens combining targeted therapies and other conventional or novel treatment strategies might serve to improve the clinical outcomes of patients with NSCLC (7).

The emergence of epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs) represented by gefitinib provides a new option for the treatment of advanced NSCLC. However, the response rate of EGFR-TKI in advanced NSCLC remains moderate. In the FLAURA trial, the progression-free survival (PFS) was only 10.2 months in patients with EGFR mutation-positive (exon 19 deletion or L858R) advanced NSCLC who received standard EGFR-TKI as first-line treatment (8). In another Phase 3 study, the PFS of gefitinib alone in newly diagnosed metastatic NSCLC with EGFR mutation was only 11.9 months with the objective response rate (ORR) being 67%, while in the combination group where patients were treated with gefitinib combined with carboplatin plus pemetrexed, ORR was 84% and PFS nearly doubled (20.9 months) (9). Patients treated with gefitinib will inevitably develop resistance, making it an unsustainable regimen (10). Moreover, rebiopsy of tumor tissue demonstrated that EGFR-TKI administration was related to changes in the tumor microenvironment, resulting in a reduced proportion of anti-tumoral CD8+ and FOSP3+ tumor-infiltrating lymphocytes, increased PD-L1 expression, and tumor mutation burden (11). This report reviews the prospects of subsequent immunotherapy after EGFR-TKI treatment, and the necessity of the combined application of EGFR-TKI and other drugs, which not only gives full play to the respective advantages of drugs but also presents potential synergistic effects against refractory malignancies. Therefore, the combination therapy has achieved good effects and has attracted much attention (See Figure 1 ).

EGFR is the protein encoded by the proto-oncogene c-erbB1, which is a member of the EGFR receptor family and is highly expressed in lung cancer. The tyrosine kinase is an essential subunit of EGFR. The transmembrane protein EGFR consists of three parts, namely the extracellular domain containing ligand binding sites, the single-transmembrane hydrophobic α-helix region, and the receptor tyrosine kinase (RTK) active intracellular domain (12–14). In the inactivated/dormant state, EGFR exists as a monomer on the cell membrane, RTK exists as a monomer and is inactive. Once the signal molecule binds to the extracellular domain of the receptor, the two form a dimer on the membrane, making the tails of the intracellular domain contact each other, thereby activating protein kinases and phosphorylating them to form signaling complexes. The extracellular region of EGFR binds to extracellular ligands to form dimers of EGFR monomers and other EGFR molecules or other ErbB family protein receptors. After dimerization, the intracellular tyrosine kinase region of EGFR binds to a molecule of ATP, which activates this region after phosphorylation. Subsequently, specific tyrosine residues on the C-terminus are phosphorylated through autophosphorylation or transphosphorylation. These phosphorylated sites provide binding sites for other downstream signaling molecules and activate the downstream signaling pathways RAS-RAF-MEK-ERK and RAS-PI3K-PTEN-AKT-mTOR, thereby promoting tumor cell proliferation, differentiation, and migration (15–18). This signal complex activates a series of intracellular biochemical reactions via RAS/RAF/ERK/MAPK pathway and PI3K/PKB/AKT pathway, triggering transcription of genes regulating the growth and differentiation of target cells (19).

The first generation EGFR-TKIs, represented by gefitinib and erlotinib, can compete for the adenosine triphosphate (ATP) binding site on the catalytic region of EGFR-TKI, inhibit RTK activity, and block the phosphorylation of RTK itself and its substrates. In turn, it blocks the conduction of downstream RAS/RAF/ERK/MAPK signaling pathways and inhibits the growth, proliferation and differentiation of tumor cells (20). The efficacy of erlotinib is slightly better than that of gefitinib, especially in patients with brain metastases from lung cancer, as a result of smaller molecular weight and better ability to cross the blood-brain barrier (21). Second-generation EGFR-TKIs Afatinib (BIBW2992) and dacomitinib (PF00299804) are oral ErbB family blockers that inhibit EGFR signaling by irreversibly binding to the ATP domain of the ErbB family kinase domain access (22, 23). A Phase 3 clinical trial demonstrated superior efficacy of dacomitinib as the first-line treatment for patients with EGFR mutation-positive NSCLC: the median PFS was 14.7 months in the dacomitinib group and 9.2 months in the gefitinib group, suggesting a better therapeutic efficacy as a second-generation EGFR-TKI (24). The third-generation EGFR-TKIs, represented by osimertinib, mainly form a covalent bond through an unsaturated acryl chain, targeting the cysteine-797 residue of the ATP binding site, and irreversibly interact with the catalytic active center of the EGFR receptor. The covalent binding of EGFR inhibits both the phosphorylation of EGFR kinase and the phosphorylation of downstream signaling substrates AKT and ERK, selectively and irreversibly targeting EGFR-sensitive mutations and T790M resistance mutations in patients with NSCLC. Osimertinib yielded promising clinical outcomes among patients with NSCLC harboring EGFR T790M mutation (25).

Studies have shown that EGFR-TKI monotherapy is safe and effective in EGFR mutation-positive NSCLC (mainly in exons 19 and 21), but not as effective in EGFR mutation-negative NSCLC (26). For patients with EGFR mutation-negative NSCLC, the recommended treatment still centers on chemotherapy, and a meta-analysis showed significant increase in PFS and ORR, but not in OS. This study suggested that chemotherapy remains a better treatment option than EGFR-TKI for patients with EGFR mutation-negative NSCLC. Docetaxel and pemetrexed are standard second- and third-line treatment options for patients with EGFR mutation-negative NSCLC. Erlotinib can also be used as a second- and third-line treatment option for such patients, especially for those with poor ECOG PS scores who cannot tolerate the side effects of chemotherapy.

Small-molecule drugs targeting the RTK is a breakthrough in the treatment of NSCLC, despite that the efficacy of monotherapy is not yet convincing. Additionally, acquired drug resistance becomes prominent with prolonged treatment, greatly affecting drug efficacy. EGFR-TKI resistance includes primary resistance and secondary resistance. Patients who do not respond to EGFR-TKIs or respond for a very brief period of time (typically less than 3 months) are considered primary resistance (27). The underlying mechanism of primary resistance include lack of target dependency, molecular alterations of genes from downstream pathways, and in some cases, incomplete drug impact caused by pharmacokinetic therapy failures (28–30).

The effects of EGFR-TKIs vary among different EGFR mutation sites, which may contribute to primary resistance to EGFR-TKI treatment. Among them, exon 19 deletion mutation and exon 21 L858R point mutation are common sensitive mutations, while exon 20 insertion or repeat mutation is associated with EGFR-TKI resistance (31). Patients who develop resistance after achieving an initial response or stable disease are considered secondary/acquired resistance, which often occurs within a year from the initiation of first- and second- generation EGFR-TKIs, and about 19 months for third generation EGFR-TKIs in the first-line setting (8). The underlying mechanism of acquired resistance can be divided into point mutations in target genes, activation of alternative signaling pathways, and phenotypic or histological transformation. Point mutations of target genes and activation of alternative signaling pathways mainly include T90M, HER2 amplification, MET amplification, BRAF amplification, PIK3CA mutation. As a receptor tyrosine kinase. although HER2 amplification is found in only 1-3% of lung adenocarcinomas, it is the second common mechanism of acquired resistance to EGFR-TKIs, occurring in 12% of cases (32, 33). The MET gene encodes c-met, which is a receptor for hepatocyte growth factor (HGF), and the binding of HGF with c-met activates the phosphorylation of HER3 (a member of the EGFR family) and the subsequent phosphorylation of the PI3K-AKT signaling pathway, one of the downstream effector of EGFR activation (30, 34). Thus, MET amplification causes acquired resistance to EGFR-TKIs and promotes the development of NSCLC. Inhibition of one signaling pathway may activate alternative signaling pathways, leading to mutual substitution between signal transduction pathways, resulting in the acquisition of resistance to EGFR-TKIs. There is mutual interference between EGFR and IGF1R due to EGFR and insulin-like growth factor 1 receptor (IGF1R)-mediated activation of downstream MAPK and AKT signaling. Inhibition of EGFR subsequently activates the IGF1R alternative signaling pathway, leading acquired resistance to afatinib as well (35, 36). Collectively, though EGFR-TKIs demonstrate promising therapeutic effects, considering drug-resistant mutations and activation of alternative signaling pathways, combination therapies are greatly necessary to provides resolutions for drug resistance and new insights into the treatment of NSCLC.

In addition to the EGFR-TKI combination therapies discussed above, alternative treatment strategies, such as nanomedicine, are emerging as potential promising therapies for lung cancer. Biodegradable nanoparticles (<1 μm) with anticancer drugs encapsulated represent a novel approach which enables effective targeted drug delivery without excess toxicities to normal cells. Anticancer drugs are either adsorbed to the surface of the nanoparticle, or encapsulated in a polymer-based cavity, and delivered to the targeted tissue either through intratumoral injection or inhaler (73). In an in vivo study, a thermosensitive nanoparticle delivery system with cisplatin and paclitaxel loaded has demonstrated good efficacy with acceptable safety as in situ treatment in a lung cancer mouse xenograft model, which suggests the potential of combining nanoparticle technology with conventional chemotherapy in the treatment of lung cancer (74).

In addition to delivery of conventional anti-cancer drugs, nanoparticles can also be used to deliver non-coding RNAs (ncRNAs) to tackle EGFR-TKI resistance. Dysregulated ncRNA have been shown to play a regulatory role in EGFR‐TKI resistance in lung cancer through modulating multiple signaling pathways, including the PI3K/AKT/mTOR, JAK/STAT, Ras/Raf/MEK/ERK signaling pathways. Therefore, ncRNAs have been explored as a potential therapeutic target for overcoming EGFR-TKI-resistance in lung cancer (75). In an in vitro study, nanoparticles encapsulating doxorubicin and short-interfering RNA were able to silence multi-drug resistance genes in lung cancer, highlighting the potential of combining nanomedicine with ncRNA in overcoming resistance to targeted therapy (76). Another study using nanoparticles to deliver short-interfering RNA against an established oncogenic long noncoding RNA (differentiation antagonizing noncoding RNA) achieved over 90% silencing, which resulted in reduced cell migration and invasion, and improved EGFR inhibitor response (77). Although still in early stage, these promising results shed light on the potential of RNA interference through nanoparticle delivery system as a novel approach for lung cancer treatment.

Although EGFR-TKI combination treatment for patients with advanced NSCLC with EGFR-sensitive mutations is extensively studied, so far, no accurate and effective prognostic biomarker has been established in this setting, which represents an unmet clinical need. In addition to exploring prognostic new biomarkers, studies suggest that inflammation, immune response, and tumor immune microenvironment are associated with the prognosis of patients with various malignancies (55, 78–81). Increased research is devoted to finding biomarkers for the prognosis of patients with NSCLC, and multiple markers have been discovered to predict disease progression status and patient prognosis from different perspectives. These include serum biomarkers C-reactive protein (CRP), inflammatory markers neutrophil to lymphocyte ratio (NLR) and lymphocyte to monocyte ratio (LMR), and nutritional level markers prognostic nutritional index (PNI), etc.

The level of CRP secreted by the liver increases rapidly in response to trauma, inflammation, and infection, and also decreases rapidly as the condition subsides. A retrospective study on patients with EGFR-sensitive mutant advanced NSCLC treated with EGFR-TKI found that, compared with patients with low pre-treatment serum CRP levels, patients with high pre-treatment serum CRP levels had a significantly poorer response to EGFR-TKI (65, 81).The results suggest that pre-treatment serum CRP levels can independently predict the prognosis of patients with advanced NSCLC who are treated with EGFR-TKIs.

The inflammatory index marker NLR can be used to reflect the body’s inflammation. When the peripheral blood neutrophil count is high or the lymphocyte count is low, the ratio becomes unbalanced, indicating a high inflammatory response. High levels of neutrophils will affect the anti-tumor effects of lymphocytes, while low lymphocyte level is not sufficient for the immune system to clear tumor cells. The above two factors can promote tumor growth. Gui Nan and Fausto Meriggi found lower NLR levels are associated with significantly longer PFS and OS in patients with NSCLC who received EGFR-TKI (82, 83). Therefore, NLR level prior to initial treatment has been explored as an independent factor of survival and prognosis in patients with NSCLC who received EGFR-TKI. The LMR is simple measurement and objective indicator of inflammation, and is an established independent prognostic factor in patients with multiple malignancies (84). A retrospective study found that high LMR levels are significantly related to the better prognosis of patients with advanced EGFR mutation-positive NSCLC treated with EGFR-TKI (78, 79). Therefore, LMR can be used to as a prognostic indicator for EGFR-TKI treatment, but the specific cut-off value for stratifying high or low LMR needs to be determined by further research.

PNI is an index that can comprehensively reflect the inflammation and nutritional status of patients with cancer (80). It is reported that PNI is related to tumor progression hence is of prognostic value in a variety of malignant tumors such as digestive tract tumors, pancreatic cancer and lung cancer (84–89). Research by Jin and colleagues found that the lower PNI is associated with poor prognosis in patients with EGFR-sensitive mutant NSCLC who received EGFR-TKI (90). Therefore, it might be worth applying the PNI prior to initiating EGFR-TKI treatment as routine clinical practice.

In addition, new biomarkers are needed to differentiate patients with EGFR mutation-positive lung cancer who are resistant to EGFR-TKIs. Although patients with EGFR mutation-positive NSCLC are commonly sensitive to EGFR-TKIs initially, the development of acquired resistance is inevitable (38). The mechanisms of acquired resistance to EGFR-TKIs can be classified into point mutations in target genes, activation of alternative signaling pathways, and phenotypic or histological transformation. T790M is a point mutation at position 790 on exon 20 of the EGFR kinase domain, where threonine is replaced by methionine, which subsequently activates downstream signaling pathways (91, 92). Residue 790 that controls entry of EGFR-TKIs into the ATP-binding pocket will increase their affinity for ATP and cause acquired resistance to targeted drugs when turned off. Detection of drug-resistant mutations by gene profile screening and by comprehensive evaluation based on system examination could potentially provide improvement to the efficacy of EGFR-TKIs in patients with NSCLC.

By now, most EGFR-TKIs aim to target the biological functions of EGFR caused by kinase activation, and to induce cytotoxicity of cancer cells, yet drug-resistant mutations, together with changes in the stability of EGFR still limits the use and efficacy of EGFR-TKIs in clinical practice. The first-generation of EFGR-TKI such as gefitinib proved effective in the most common 19del and L858R mutations, but not in resistant mutations such as T790M, which accounts for 60% of drug-resistant EGFR mutations (93). New EGFR-TKIs represented by osimertinib and afatinib are promising solutions to the resistant mutations. Moreover, monoclonal antibodies targeting EGFR and its downstream signaling pathways were reported in treating colorectal cancer, head and neck carcinomas, but few reports comprehensively investigated the clinical outcomes in NSCLC (94). This review discussed EGFR-TKI resistance mechanisms and its clinical implications in NSCLC, summarized current EGFR-TKI-based combination therapy regimens and overviewed novel treatment approaches including nanoparticles encapsulating non-coding RNAs for overcoming EGFR-TKI resistance in NSCLC. In summary, EGFR-TKIs combination therapies are potent treatment strategies against NSCLC, and further research are needed to overcome resistance and other barriers that hinder their clinical application.

WP, GZ, and YF provided the direction and guidance of this manuscript. CY and QP wrote the whole manuscript. JY, BS, and ZZ made significant revisions to the manuscript. All authors contributed to the article and approved the submitted version.